Psilocybin 'Homological Scaffold' Connectivity - Recent Study

[nAON]

Some of you may recognise this visual that has, for me at least, been rinsing its way through my facebook for the last few weeks, apparently showing that psilocybin increases connectivity and activity across discrete brain regions. [source = http://rsif.royalsocietypublishing.org/content/11/101/20140873.full.pdf+html]

I have attempted to read the paper and failed miserably, too many equations and network-lingo for me. Could anyone here help explain this to me? How was this visual created, what does it show exactly, and what are the implications? How does it integrate into previous fMRI findings that show overall decreased connectivity in the brain on psilocybin?

 

[sekio]

Simplified visualization of the persistence homological scaffolds. The persistence homological scaffolds Graphic (a) and Graphic (b) are shown for comparison. For ease of visualization, only the links heavier than 80 (the weight at which the distributions in figure 5a bifurcate) are shown. This value is slightly smaller than the bifurcation point of the weights distributions in figure 5a. In both networks, colours represent communities obtained by modularity [49] optimization on the placebo persistence scaffold using the Louvain method [50] and are used to show the departure of the psilocybin connectivity structure from the placebo baseline. The width of the links is proportional to their weight and the size of the nodes is proportional to their strength. Note that the proportion of heavy links between communities is much higher (and very different) in the psilocybin group, suggesting greater integration. A labelled version of the two scaffolds is available as GEXF graph files as the electronic supplementary material.

The two key results of the analysis of the homological scaffolds can therefore be summarized as follows (i) there is an increased integration between cortical regions in the psilocybin state and (ii) this integration is supported by a persistent scaffold of a set of edges that support cross modular connectivity probably as a result of the stimulation of the 5HT2A receptors in the cortex.

My graph theory is somewhat lacking but I think, very basically, it's a diagram of associations between fMRI signals of various regions of the brain. I guess it shows that regional brain activity is more highly synchronized in psilocybin volunteers. It's a damn low dose of psilocybin though (2mg).

Facebook is probably about the least qualified place to discuss this, anyway. I'm sure a lot of people interpret this as "mushroomz makes u smart lol".

 

[nAON]

I wonder if there's any interesting (and more importantly, colourful!) EEG data on this then?

 

[Zilpe]

Interesting, I know someone who was using homology to study fmri data. Unfortunately I was not really paying attention during her presentation so I don't really have anything to contribute. I'll try to read the paper later to see if I can figure out what's going on. For those interested homology is the primary tool used in algebraic topology, the study of algebraic structures assigned to topological spaces.

 

[specialspack]

I'll do my best with this...

I have attempted to read the paper and failed miserably, too many equations and network-lingo for me. Could anyone here help explain this to me? How was this visual created, what does it show exactly, and what are the implications? How does it integrate into previous fMRI findings that show overall decreased connectivity in the brain on psilocybin?

How was the visual created? I'm afraid I can't really shed much light here - the methods are described in the paper for those of you with network & graph theory backgrounds. The different coloured regions represent modules in the network - it's my understanding that these correspond loosely to resting state functional networks. Although I'm not clear how good the mapping is and what colour corresponds to what module. Thus, in the placebo state, the picture is dominated by strong, heavily weighted intra-network links, with a few inter-network connections spanning the diagram. In contrast, the psilocybin state there are many more heavily weighted inter-network links.

How does this relate to previous fMRI psilocybin findings?

First it's important to remember that all of these papers are from the same dataset of 15 subjects. The various papers published so far don't show overall decreased connectivity. They showed a) there were only net decreases in BOLD activity b) there were decreases in within-network functional connectivity within the default mode network (DMN) and dorsal attention network (DAN) c) there were increases in between-network connectivity between the DMN and various task positive networks (using a seed based analysis) d) there were many increases in between-network connectivity (using an ICA based analysis to determine multiple networks), along with a smaller number of decreases (mainly between occipital visual networks and other regions).

I guess it shows that regional brain activity is more highly synchronized in psilocybin volunteers. It's a damn low dose of psilocybin though (2mg).

NB: 2mg is an IV dose. This compares objectively (using the 5D-ASC) to a 15mg oral dose. Subjects are definitely having a psychedelic experience (although the time course is massively truncated and is similar to smoked DMT).

The overall picture here is that psilocybin disrupts within-network connectivity, and increases between-network connectivity, moving the brain away from a relatively constrained state to a more chaotic one. The Frontiers paper explores this hypothesis at length:
http://journal.frontiersin.org/Journal/10.3389/fnhum.2014.00020/abstract

The same group has published a MEG study - http://www.jneurosci.org/content/33/38/15171.short - doesn't look at functional connectivity though, just decreases in power globally in within networks.

I think the only EEG psilocybin work done by the Zurich lab is task based; no one has done any resting state EEG that I'm aware of.

 

[serotonin2A]

I don't think the dose comparison (2 mg IV = 15 mg po) is correct. IIRC, they used the APZ rating scale to assess the intensity of the psychedelic state, but they only reported data for a few individual questions, not the entire test. The APZ isn't supposed to work like that, it has only been validated using the global score or the subscales (groups of questions that are supposedly relatively homogeneous). You can't just pick and choose questions to use for a study, and then try to compare the data to previous results ontained using the full test. So, I don't think they have really published a full characterization of what 2 mg IV psilocybin is doing.

 

[ParappaTheRapper]

The patterns remind me of visual observances of eye movrment recording to pictures ive studied in psychology.

 

[specialspack]

I don't think the dose comparison (2 mg IV = 15 mg po) is correct. IIRC, they used the APZ rating scale to assess the intensity of the psychedelic state, but they only reported data for a few individual questions, not the entire test. The APZ isn't supposed to work like that, it has only been validated using the global score or the subscales (groups of questions that are supposedly relatively homogeneous). You can't just pick and choose questions to use for a study, and then try to compare the data to previous results ontained using the full test. So, I don't think they have really published a full characterization of what 2 mg IV psilocybin is doing.

Carhart-Harris et al 2011
"The administration of psilocybin to healthy, hallucinogen-experienced volunteers in a mock-functional magnetic resonance imaging environment: a preliminary investigation of tolerability"

Over the next 60–90min subjects completed the 5-Dimensions of Altered States of Consciousness questionnaire (5D-ASC) (Dittrich 1998, translated from the original German into English by Felix Hasler and Rael Cahn), a 94-item self-rated questionnaire.

After the acute drug effects had subsided, subjects completed the 5-D ASC. Global and subscale ratings were generally lower for subjects administered 1.5 mg, but not to a statistically significant degree. At 1.5 mg, the mean global rating was roughly consistent with ratings previously recorded after a ‘low’ oral dose of psilocybin (i.e. 8 mg), and at 2mg, mean global ratings were roughly consistent with ratings recorded after a ‘medium’ oral dose of psilocybin (i.e. 15 mg) (Hasler et al., 2004). Figure 3 shows the ratings for the 94-item 5-D ASC expressed as a percentage of the maximum possible ratings for each dimension.

 

[serotonin2A]

Unfortunately, they didn't report the same data in the PNAS study. The full 5D-ASC data from the Journal of Psychopharmacology are problematic, and do not support the conclusion that the effects of 2 mg IV psilocybin are really equivalent to 15 mg PO. Looking at one of Vollenweider's publications, the magnitude of the 5D-ASC scores with IV psilocybin match those found with a medium oral psilocybin dose (0.215 mg/kg), but in the Swiss studies the baseline scores are all much much lower, generally around 1%. By contrast, in David Nutt's studies, the baseline scores sometimes reach 10-15%. Therefore, the magnitude of the increase with IV psilocybin is much lower then with PO psilocybin. It's not clear from the graph if the change induced with IV psilocybin is actually significant, but it looks like it might not be, especially given the size of the error bars.

For the 5D-ASC, it is really the change from baseline, rather then the score itself, that accurately defines the effect of a drug. The baseline scores can vary depending on the particular subjects. If a particular subject shows a 15% response to one question at baseline, and a 25% response to the same question after drug treatment, then the drug hasn't done very much. That would obviously be different from a subject who goes from 1% to 25% after receiving a drug.

(Hasler et al., Acute psychological and physiological effects of psilocybin in healthy humans: a double-blind, placebo-controlled dose–effect study. Psychopharmacology vol 172, p. 145 (2004))

 

[specialspack]

Unfortunately, they didn't report the same data in the PNAS study. The full 5D-ASC data from the Journal of Psychopharmacology are problematic, and do not support the conclusion that the effects of 2 mg IV psilocybin are really equivalent to 15 mg PO. Looking at one of Vollenweider's publications, the magnitude of the 5D-ASC scores with IV psilocybin match those found with a medium oral psilocybin dose (0.215 mg/kg), but in the Swiss studies the baseline scores are all much much lower, generally around 1%. By contrast, in David Nutt's studies, the baseline scores sometimes reach 10-15%. Therefore, the magnitude of the increase with IV psilocybin is much lower then with PO psilocybin. It's not clear from the graph if the change induced with IV psilocybin is actually significant, but it looks like it might not be, especially given the size of the error bars.

For the 5D-ASC, it is really the change from baseline, rather then the score itself, that accurately defines the effect of a drug. The baseline scores can vary depending on the particular subjects. If a particular subject shows a 15% response to one question at baseline, and a 25% response to the same question after drug treatment, then the drug hasn't done very much. That would obviously be different from a subject who goes from 1% to 25% after receiving a drug.

(Hasler et al., Acute psychological and physiological effects of psilocybin in healthy humans: a double-blind, placebo-controlled dose–effect study. Psychopharmacology vol 172, p. 145 (2004))

I agree the lack of full ASC data in the PNAS paper is not really ideal, and I wish they had collected it.

However - unless I'm missing something, in the Psychopharmacology paper, they don't present the baseline ASC scores - the graph shown is 1.5 and 2 mg psilocybin - so when you say that baseline scores sometimes reach 10-15%, I think you're referring to the 1.5mg scores?

It maybe that 15mg oral is slightly stronger than 2mg IV - what I don't think can be contested is that 2mg IV is certainly a psychedelic dose. I know Vollenweider wants to say it isn't really, as the fMRI and MEG data are contrary to their PET study.

 

[serotonin2A]

I thought the smaller bars were baseline values. I don't have the paper in front of me right now, but I don't understand why they wouldn't show baseline values, since you need them to understand how much of an effect psilocybin is having. The 5D-ASC is a relative scale not an absolute scale, so you really have to see some type of baseline, either a placebo arm or at least an assessment prior to giving the drug.

Vollenweider is one of the only people who has tested psilocybin IV and PO, so I tend to believe him when he says that IV psilocybin is not a suitable route for studies.

The problem with giving psilocybin IV for a fMRI study is that it markedly increases the intensity of the cardiovascular response, which could confound the fMRI data. In a previous paper, at 4 mg IV psilocybin, the subject completely lost consciousness. It's already bad enough that 5-HT1A agonists produce effects on fMRI.

 

[specialspack]

I thought the smaller bars were baseline values. I don't have the paper in front of me right now, but I don't understand why they wouldn't show baseline values, since you need them to understand how much of an effect psilocybin is having. The 5D-ASC is a relative scale not an absolute scale, so you really have to see some type of baseline, either a placebo arm or at least an assessment prior to giving the drug.

No the smaller bars are definitely 1.5mg. It was a dose-escalating pilot study, aimed at finding the best dose with a balance between psychedelic effects and tolerability in the fMRI scanner, hence no placebo dose was done. It's an interesting point about the ASC being relative, rather than absolute - I don't recall reading any specific points to this regard in Dittrich or Studerus papers. Out of the many trials, did any show more than a few % in the placebo groups? If not, it seems highly unlikely that the Imperial group's subjects happened to have non-normal distribution.

I mean - is the suggestion seriously that because the placebo data wasn't reported, therefore 2mg IV is non-psychdelic..?

Vollenweider is one of the only people who has tested psilocybin IV and PO, so I tend to believe him when he says that IV psilocybin is not a suitable route for studies.

I wasn't aware the Zurich lab had done any IV studies - are they published anywhere?

I think it's important to recognize that the Zurich and Imperial teams have a degree of rivalry - and thus tend to be quite dismissive of each others research! I don't regard either of them as dispassionate commentators, and they have vested interests in pushing their particular claims. Vollenweider has been very dismissive of the Imperial work, and Carhart-Harris has been equally critical of the Zurich PET studies. Also Vollenweider tends to be very critical of just about everyone else in the field! His comments in this recent Science article are pretty illustrative.

The problem with giving psilocybin IV for a fMRI study is that it markedly increases the intensity of the cardiovascular response, which could confound the fMRI data. In a previous paper, at 4 mg IV psilocybin, the subject completely lost consciousness. It's already bad enough that 5-HT1A agonists produce effects on fMRI.

Again - interesting - where's the 4mg study published?
The fact of psilocybin IV physiological effect is a separate issue from debating its psychedelic effect, which is what we were discussing. The PNAS paper used physiological data as regressors in the general linear model used for the fMRI analysis, and saw no significant changes (in the supplementary data). Also the subsequent MEG paper supports the idea that it's changes in neuronal activity rather than just CBF related.

If you have refs for 5HT1A agonist fMRI send em over - but surely what we're interested in here is the psychedelic state - whether or not 5HT1A contributes to that effect or not? We'll have to wait for fMRI of the Cimbi tracers... 25B- and 25CN-NBOMe - which might not be too far off!

I certainly don't think the Imperial work is beyond criticism, but trying to deny the psychedelic nature of IV psilocybin seems a bit daft - when there are plenty of methodological angles of attack you could take. For instance, recent advances in motion correction...

 

[serotonin2A]

I'm critical of their work for the simple reason that there is a flaw in it. The 5D-ASC is a relative scale, not an absolute scale, so it is not really possible to interpret the results without the baseline data. I don't want to make the assumption that 2 mg I've psilocybin is psychedelic--that's what the 5D-ASC is supposed to show, and I can't make that judgement without seeing the data. The fact that they could have easily included the baseline data, but didn't, makes me think that maybe they left it out on purpose. Do you honestly believe that I shouldn't expect to see their 5D-ASC data reported in a way that can actually be interpreted and compared with data from other groups?

What you said about Vollenwider doesn't mean that he isn't right about IV psilocybin being an inferior route of administration. The psilocybin metabolism study that Vollenweider ran with Felix Hasler describes the effects of 2 and 4 mg IV psilocybin.

It was my impression that the MEG data was consistent with the interpretation that PET and fMRI are measuring two different effects of psilocybin (PET measuring firing activity and fMRI measuring oscillatory power). IIRC, they modeled the MEG findings and showed that they were caused by an increase in firing of layer V pyramidal neurons in PFC, which is more consistant with the PET data. Artigas recently published a study in rats with 5-MeO-DMT, and the results seemed to match up with both the PET data and the fMRI data. 5-MeO-DMT reduced BOLD signal in PFC, but also increased neuronal firing in PFC.